Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,202 @@
+---
+base_model: vidore/colqwen2-base
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "vidore/colqwen2-base",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc856312174dc99a4c7f88a2c54d9590a3b3f5b5a86e2728d7138c7f4758c6d5
+size 74018232

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

checkpoint-2310/README.md

@@ -0,0 +1,202 @@
+---
+base_model: ./models/colqwen2_base
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.11.1

checkpoint-2310/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "./models/colqwen2_base",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-2310/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc856312174dc99a4c7f88a2c54d9590a3b3f5b5a86e2728d7138c7f4758c6d5
+size 74018232

checkpoint-2310/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5fc33e4e0f694908053e47d9c32ed593fa555801312d8d9d46b3a748b41bba0
+size 148262384

checkpoint-2310/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b34c39407e5982e3760c227b18a85561bb7fb0c385276cbb7370cab74802ef2f
+size 14244

checkpoint-2310/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e687ca3f782277f406bcba194795e34522524b1c24163a5c1c24714bf5f94a0d
+size 1064

checkpoint-2310/trainer_state.json

@@ -0,0 +1,1834 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 100,
+  "global_step": 2310,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021645021645021644,
+      "grad_norm": 2.59375,
+      "learning_rate": 5e-05,
+      "loss": 0.9442,
+      "step": 10
+    },
+    {
+      "epoch": 0.04329004329004329,
+      "grad_norm": 1.109375,
+      "learning_rate": 0.0001,
+      "loss": 0.6729,
+      "step": 20
+    },
+    {
+      "epoch": 0.06493506493506493,
+      "grad_norm": 0.765625,
+      "learning_rate": 0.00015,
+      "loss": 0.5235,
+      "step": 30
+    },
+    {
+      "epoch": 0.08658008658008658,
+      "grad_norm": 0.87890625,
+      "learning_rate": 0.0002,
+      "loss": 0.3883,
+      "step": 40
+    },
+    {
+      "epoch": 0.10822510822510822,
+      "grad_norm": 0.8203125,
+      "learning_rate": 0.00025,
+      "loss": 0.3269,
+      "step": 50
+    },
+    {
+      "epoch": 0.12987012987012986,
+      "grad_norm": 0.5,
+      "learning_rate": 0.0003,
+      "loss": 0.2779,
+      "step": 60
+    },
+    {
+      "epoch": 0.15151515151515152,
+      "grad_norm": 0.54296875,
+      "learning_rate": 0.00035,
+      "loss": 0.2607,
+      "step": 70
+    },
+    {
+      "epoch": 0.17316017316017315,
+      "grad_norm": 0.78515625,
+      "learning_rate": 0.0004,
+      "loss": 0.246,
+      "step": 80
+    },
+    {
+      "epoch": 0.19480519480519481,
+      "grad_norm": 0.474609375,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 0.2469,
+      "step": 90
+    },
+    {
+      "epoch": 0.21645021645021645,
+      "grad_norm": 0.55078125,
+      "learning_rate": 0.0005,
+      "loss": 0.2327,
+      "step": 100
+    },
+    {
+      "epoch": 0.21645021645021645,
+      "eval_loss": 0.1517176628112793,
+      "eval_runtime": 23.1527,
+      "eval_samples_per_second": 21.596,
+      "eval_steps_per_second": 0.691,
+      "step": 100
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 0.51953125,
+      "learning_rate": 0.000497737556561086,
+      "loss": 0.231,
+      "step": 110
+    },
+    {
+      "epoch": 0.2597402597402597,
+      "grad_norm": 0.46875,
+      "learning_rate": 0.000495475113122172,
+      "loss": 0.214,
+      "step": 120
+    },
+    {
+      "epoch": 0.2813852813852814,
+      "grad_norm": 0.4453125,
+      "learning_rate": 0.000493212669683258,
+      "loss": 0.2164,
+      "step": 130
+    },
+    {
+      "epoch": 0.30303030303030304,
+      "grad_norm": 0.470703125,
+      "learning_rate": 0.0004909502262443439,
+      "loss": 0.2327,
+      "step": 140
+    },
+    {
+      "epoch": 0.3246753246753247,
+      "grad_norm": 0.392578125,
+      "learning_rate": 0.0004886877828054299,
+      "loss": 0.2104,
+      "step": 150
+    },
+    {
+      "epoch": 0.3463203463203463,
+      "grad_norm": 0.4765625,
+      "learning_rate": 0.00048642533936651587,
+      "loss": 0.2249,
+      "step": 160
+    },
+    {
+      "epoch": 0.36796536796536794,
+      "grad_norm": 0.3125,
+      "learning_rate": 0.0004841628959276018,
+      "loss": 0.2179,
+      "step": 170
+    },
+    {
+      "epoch": 0.38961038961038963,
+      "grad_norm": 0.453125,
+      "learning_rate": 0.0004819004524886878,
+      "loss": 0.2174,
+      "step": 180
+    },
+    {
+      "epoch": 0.41125541125541126,
+      "grad_norm": 0.5390625,
+      "learning_rate": 0.0004796380090497738,
+      "loss": 0.2159,
+      "step": 190
+    },
+    {
+      "epoch": 0.4329004329004329,
+      "grad_norm": 0.392578125,
+      "learning_rate": 0.0004773755656108598,
+      "loss": 0.2166,
+      "step": 200
+    },
+    {
+      "epoch": 0.4329004329004329,
+      "eval_loss": 0.13433273136615753,
+      "eval_runtime": 18.3553,
+      "eval_samples_per_second": 27.24,
+      "eval_steps_per_second": 0.872,
+      "step": 200
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.44921875,
+      "learning_rate": 0.00047511312217194567,
+      "loss": 0.2232,
+      "step": 210
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.447265625,
+      "learning_rate": 0.00047285067873303167,
+      "loss": 0.2034,
+      "step": 220
+    },
+    {
+      "epoch": 0.49783549783549785,
+      "grad_norm": 0.36328125,
+      "learning_rate": 0.00047058823529411766,
+      "loss": 0.2222,
+      "step": 230
+    },
+    {
+      "epoch": 0.5194805194805194,
+      "grad_norm": 0.68359375,
+      "learning_rate": 0.00046832579185520365,
+      "loss": 0.2043,
+      "step": 240
+    },
+    {
+      "epoch": 0.5411255411255411,
+      "grad_norm": 0.431640625,
+      "learning_rate": 0.0004660633484162896,
+      "loss": 0.2054,
+      "step": 250
+    },
+    {
+      "epoch": 0.5627705627705628,
+      "grad_norm": 0.38671875,
+      "learning_rate": 0.0004638009049773756,
+      "loss": 0.1922,
+      "step": 260
+    },
+    {
+      "epoch": 0.5844155844155844,
+      "grad_norm": 0.349609375,
+      "learning_rate": 0.0004615384615384616,
+      "loss": 0.196,
+      "step": 270
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 0.37109375,
+      "learning_rate": 0.0004592760180995475,
+      "loss": 0.2008,
+      "step": 280
+    },
+    {
+      "epoch": 0.6277056277056277,
+      "grad_norm": 0.431640625,
+      "learning_rate": 0.00045701357466063346,
+      "loss": 0.1983,
+      "step": 290
+    },
+    {
+      "epoch": 0.6493506493506493,
+      "grad_norm": 0.4609375,
+      "learning_rate": 0.00045475113122171945,
+      "loss": 0.1964,
+      "step": 300
+    },
+    {
+      "epoch": 0.6493506493506493,
+      "eval_loss": 0.12437203526496887,
+      "eval_runtime": 19.6174,
+      "eval_samples_per_second": 25.488,
+      "eval_steps_per_second": 0.816,
+      "step": 300
+    },
+    {
+      "epoch": 0.670995670995671,
+      "grad_norm": 0.369140625,
+      "learning_rate": 0.00045248868778280545,
+      "loss": 0.1922,
+      "step": 310
+    },
+    {
+      "epoch": 0.6926406926406926,
+      "grad_norm": 0.3671875,
+      "learning_rate": 0.00045022624434389144,
+      "loss": 0.1912,
+      "step": 320
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.435546875,
+      "learning_rate": 0.0004479638009049774,
+      "loss": 0.1927,
+      "step": 330
+    },
+    {
+      "epoch": 0.7359307359307359,
+      "grad_norm": 0.375,
+      "learning_rate": 0.0004457013574660634,
+      "loss": 0.1919,
+      "step": 340
+    },
+    {
+      "epoch": 0.7575757575757576,
+      "grad_norm": 0.365234375,
+      "learning_rate": 0.0004434389140271493,
+      "loss": 0.1941,
+      "step": 350
+    },
+    {
+      "epoch": 0.7792207792207793,
+      "grad_norm": 0.50390625,
+      "learning_rate": 0.0004411764705882353,
+      "loss": 0.1939,
+      "step": 360
+    },
+    {
+      "epoch": 0.8008658008658008,
+      "grad_norm": 0.9765625,
+      "learning_rate": 0.00043891402714932125,
+      "loss": 0.2017,
+      "step": 370
+    },
+    {
+      "epoch": 0.8225108225108225,
+      "grad_norm": 0.240234375,
+      "learning_rate": 0.00043665158371040724,
+      "loss": 0.2463,
+      "step": 380
+    },
+    {
+      "epoch": 0.8441558441558441,
+      "grad_norm": 0.322265625,
+      "learning_rate": 0.00043438914027149324,
+      "loss": 0.2453,
+      "step": 390
+    },
+    {
+      "epoch": 0.8658008658008658,
+      "grad_norm": 0.26953125,
+      "learning_rate": 0.00043212669683257923,
+      "loss": 0.2457,
+      "step": 400
+    },
+    {
+      "epoch": 0.8658008658008658,
+      "eval_loss": 0.15143588185310364,
+      "eval_runtime": 20.4597,
+      "eval_samples_per_second": 24.438,
+      "eval_steps_per_second": 0.782,
+      "step": 400
+    },
+    {
+      "epoch": 0.8874458874458875,
+      "grad_norm": 0.28515625,
+      "learning_rate": 0.00042986425339366517,
+      "loss": 0.228,
+      "step": 410
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.2177734375,
+      "learning_rate": 0.0004276018099547511,
+      "loss": 0.2139,
+      "step": 420
+    },
+    {
+      "epoch": 0.9307359307359307,
+      "grad_norm": 0.271484375,
+      "learning_rate": 0.0004253393665158371,
+      "loss": 0.2224,
+      "step": 430
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.375,
+      "learning_rate": 0.0004230769230769231,
+      "loss": 0.2108,
+      "step": 440
+    },
+    {
+      "epoch": 0.974025974025974,
+      "grad_norm": 0.322265625,
+      "learning_rate": 0.00042081447963800904,
+      "loss": 0.2003,
+      "step": 450
+    },
+    {
+      "epoch": 0.9956709956709957,
+      "grad_norm": 0.47265625,
+      "learning_rate": 0.00041855203619909503,
+      "loss": 0.1827,
+      "step": 460
+    },
+    {
+      "epoch": 1.0173160173160174,
+      "grad_norm": 0.53515625,
+      "learning_rate": 0.000416289592760181,
+      "loss": 0.1791,
+      "step": 470
+    },
+    {
+      "epoch": 1.0389610389610389,
+      "grad_norm": 1.84375,
+      "learning_rate": 0.00041402714932126697,
+      "loss": 0.1836,
+      "step": 480
+    },
+    {
+      "epoch": 1.0606060606060606,
+      "grad_norm": 0.36328125,
+      "learning_rate": 0.0004117647058823529,
+      "loss": 0.1804,
+      "step": 490
+    },
+    {
+      "epoch": 1.0822510822510822,
+      "grad_norm": 0.375,
+      "learning_rate": 0.0004095022624434389,
+      "loss": 0.1977,
+      "step": 500
+    },
+    {
+      "epoch": 1.0822510822510822,
+      "eval_loss": 0.1261492371559143,
+      "eval_runtime": 20.0605,
+      "eval_samples_per_second": 24.925,
+      "eval_steps_per_second": 0.798,
+      "step": 500
+    },
+    {
+      "epoch": 1.103896103896104,
+      "grad_norm": 32.5,
+      "learning_rate": 0.0004072398190045249,
+      "loss": 0.7592,
+      "step": 510
+    },
+    {
+      "epoch": 1.1255411255411256,
+      "grad_norm": 1.0703125,
+      "learning_rate": 0.0004049773755656109,
+      "loss": 0.614,
+      "step": 520
+    },
+    {
+      "epoch": 1.1471861471861473,
+      "grad_norm": 3.71875,
+      "learning_rate": 0.0004027149321266968,
+      "loss": 0.2444,
+      "step": 530
+    },
+    {
+      "epoch": 1.1688311688311688,
+      "grad_norm": 1.2421875,
+      "learning_rate": 0.0004004524886877828,
+      "loss": 0.1821,
+      "step": 540
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.5546875,
+      "learning_rate": 0.00039819004524886876,
+      "loss": 0.1811,
+      "step": 550
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 0.373046875,
+      "learning_rate": 0.00039592760180995475,
+      "loss": 0.1717,
+      "step": 560
+    },
+    {
+      "epoch": 1.2337662337662338,
+      "grad_norm": 0.58984375,
+      "learning_rate": 0.00039366515837104075,
+      "loss": 0.1725,
+      "step": 570
+    },
+    {
+      "epoch": 1.2554112554112553,
+      "grad_norm": 0.298828125,
+      "learning_rate": 0.0003914027149321267,
+      "loss": 0.1713,
+      "step": 580
+    },
+    {
+      "epoch": 1.277056277056277,
+      "grad_norm": 0.58203125,
+      "learning_rate": 0.0003891402714932127,
+      "loss": 0.1655,
+      "step": 590
+    },
+    {
+      "epoch": 1.2987012987012987,
+      "grad_norm": 0.8046875,
+      "learning_rate": 0.0003868778280542987,
+      "loss": 0.1835,
+      "step": 600
+    },
+    {
+      "epoch": 1.2987012987012987,
+      "eval_loss": 0.12451652437448502,
+      "eval_runtime": 20.7534,
+      "eval_samples_per_second": 24.092,
+      "eval_steps_per_second": 0.771,
+      "step": 600
+    },
+    {
+      "epoch": 1.3203463203463204,
+      "grad_norm": 0.421875,
+      "learning_rate": 0.00038461538461538467,
+      "loss": 0.1626,
+      "step": 610
+    },
+    {
+      "epoch": 1.341991341991342,
+      "grad_norm": 0.439453125,
+      "learning_rate": 0.00038235294117647055,
+      "loss": 0.1713,
+      "step": 620
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.380859375,
+      "learning_rate": 0.00038009049773755655,
+      "loss": 0.161,
+      "step": 630
+    },
+    {
+      "epoch": 1.3852813852813852,
+      "grad_norm": 0.478515625,
+      "learning_rate": 0.00037782805429864254,
+      "loss": 0.1823,
+      "step": 640
+    },
+    {
+      "epoch": 1.406926406926407,
+      "grad_norm": 0.404296875,
+      "learning_rate": 0.00037556561085972854,
+      "loss": 0.1822,
+      "step": 650
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.4296875,
+      "learning_rate": 0.0003733031674208145,
+      "loss": 0.1675,
+      "step": 660
+    },
+    {
+      "epoch": 1.4502164502164503,
+      "grad_norm": 0.376953125,
+      "learning_rate": 0.00037104072398190047,
+      "loss": 0.1736,
+      "step": 670
+    },
+    {
+      "epoch": 1.4718614718614718,
+      "grad_norm": 0.412109375,
+      "learning_rate": 0.00036877828054298646,
+      "loss": 0.1614,
+      "step": 680
+    },
+    {
+      "epoch": 1.4935064935064934,
+      "grad_norm": 0.373046875,
+      "learning_rate": 0.0003665158371040724,
+      "loss": 0.1795,
+      "step": 690
+    },
+    {
+      "epoch": 1.5151515151515151,
+      "grad_norm": 0.345703125,
+      "learning_rate": 0.00036425339366515834,
+      "loss": 0.183,
+      "step": 700
+    },
+    {
+      "epoch": 1.5151515151515151,
+      "eval_loss": 0.10722990334033966,
+      "eval_runtime": 19.3953,
+      "eval_samples_per_second": 25.779,
+      "eval_steps_per_second": 0.825,
+      "step": 700
+    },
+    {
+      "epoch": 1.5367965367965368,
+      "grad_norm": 0.380859375,
+      "learning_rate": 0.00036199095022624434,
+      "loss": 0.1647,
+      "step": 710
+    },
+    {
+      "epoch": 1.5584415584415585,
+      "grad_norm": 0.369140625,
+      "learning_rate": 0.00035972850678733033,
+      "loss": 0.1649,
+      "step": 720
+    },
+    {
+      "epoch": 1.5800865800865802,
+      "grad_norm": 0.283203125,
+      "learning_rate": 0.0003574660633484163,
+      "loss": 0.1676,
+      "step": 730
+    },
+    {
+      "epoch": 1.601731601731602,
+      "grad_norm": 0.41796875,
+      "learning_rate": 0.00035520361990950226,
+      "loss": 0.1625,
+      "step": 740
+    },
+    {
+      "epoch": 1.6233766233766234,
+      "grad_norm": 0.365234375,
+      "learning_rate": 0.00035294117647058826,
+      "loss": 0.1714,
+      "step": 750
+    },
+    {
+      "epoch": 1.645021645021645,
+      "grad_norm": 0.31640625,
+      "learning_rate": 0.0003506787330316742,
+      "loss": 0.1612,
+      "step": 760
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.4921875,
+      "learning_rate": 0.0003484162895927602,
+      "loss": 0.1623,
+      "step": 770
+    },
+    {
+      "epoch": 1.6883116883116882,
+      "grad_norm": 0.380859375,
+      "learning_rate": 0.00034615384615384613,
+      "loss": 0.1508,
+      "step": 780
+    },
+    {
+      "epoch": 1.70995670995671,
+      "grad_norm": 0.34375,
+      "learning_rate": 0.0003438914027149321,
+      "loss": 0.1632,
+      "step": 790
+    },
+    {
+      "epoch": 1.7316017316017316,
+      "grad_norm": 0.34765625,
+      "learning_rate": 0.0003416289592760181,
+      "loss": 0.1682,
+      "step": 800
+    },
+    {
+      "epoch": 1.7316017316017316,
+      "eval_loss": 0.10919010639190674,
+      "eval_runtime": 27.8108,
+      "eval_samples_per_second": 17.979,
+      "eval_steps_per_second": 0.575,
+      "step": 800
+    },
+    {
+      "epoch": 1.7532467532467533,
+      "grad_norm": 0.310546875,
+      "learning_rate": 0.0003393665158371041,
+      "loss": 0.1514,
+      "step": 810
+    },
+    {
+      "epoch": 1.774891774891775,
+      "grad_norm": 0.359375,
+      "learning_rate": 0.00033710407239819005,
+      "loss": 0.1536,
+      "step": 820
+    },
+    {
+      "epoch": 1.7965367965367967,
+      "grad_norm": 0.248046875,
+      "learning_rate": 0.000334841628959276,
+      "loss": 0.1723,
+      "step": 830
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.484375,
+      "learning_rate": 0.000332579185520362,
+      "loss": 0.168,
+      "step": 840
+    },
+    {
+      "epoch": 1.8398268398268398,
+      "grad_norm": 0.35546875,
+      "learning_rate": 0.000330316742081448,
+      "loss": 0.1515,
+      "step": 850
+    },
+    {
+      "epoch": 1.8614718614718615,
+      "grad_norm": 0.298828125,
+      "learning_rate": 0.0003280542986425339,
+      "loss": 0.1694,
+      "step": 860
+    },
+    {
+      "epoch": 1.883116883116883,
+      "grad_norm": 0.3671875,
+      "learning_rate": 0.0003257918552036199,
+      "loss": 0.1679,
+      "step": 870
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.357421875,
+      "learning_rate": 0.0003235294117647059,
+      "loss": 0.1632,
+      "step": 880
+    },
+    {
+      "epoch": 1.9264069264069263,
+      "grad_norm": 0.3828125,
+      "learning_rate": 0.0003212669683257919,
+      "loss": 0.1619,
+      "step": 890
+    },
+    {
+      "epoch": 1.948051948051948,
+      "grad_norm": 0.380859375,
+      "learning_rate": 0.0003190045248868778,
+      "loss": 0.1632,
+      "step": 900
+    },
+    {
+      "epoch": 1.948051948051948,
+      "eval_loss": 0.11044134944677353,
+      "eval_runtime": 18.793,
+      "eval_samples_per_second": 26.606,
+      "eval_steps_per_second": 0.851,
+      "step": 900
+    },
+    {
+      "epoch": 1.9696969696969697,
+      "grad_norm": 0.306640625,
+      "learning_rate": 0.0003167420814479638,
+      "loss": 0.1625,
+      "step": 910
+    },
+    {
+      "epoch": 1.9913419913419914,
+      "grad_norm": 0.384765625,
+      "learning_rate": 0.0003144796380090498,
+      "loss": 0.1616,
+      "step": 920
+    },
+    {
+      "epoch": 2.012987012987013,
+      "grad_norm": 0.318359375,
+      "learning_rate": 0.00031221719457013577,
+      "loss": 0.1433,
+      "step": 930
+    },
+    {
+      "epoch": 2.034632034632035,
+      "grad_norm": 0.310546875,
+      "learning_rate": 0.0003099547511312217,
+      "loss": 0.1483,
+      "step": 940
+    },
+    {
+      "epoch": 2.0562770562770565,
+      "grad_norm": 0.322265625,
+      "learning_rate": 0.0003076923076923077,
+      "loss": 0.1398,
+      "step": 950
+    },
+    {
+      "epoch": 2.0779220779220777,
+      "grad_norm": 0.83203125,
+      "learning_rate": 0.0003054298642533937,
+      "loss": 0.1488,
+      "step": 960
+    },
+    {
+      "epoch": 2.0995670995670994,
+      "grad_norm": 0.279296875,
+      "learning_rate": 0.00030316742081447964,
+      "loss": 0.1468,
+      "step": 970
+    },
+    {
+      "epoch": 2.121212121212121,
+      "grad_norm": 0.337890625,
+      "learning_rate": 0.0003009049773755656,
+      "loss": 0.1503,
+      "step": 980
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.2373046875,
+      "learning_rate": 0.00029864253393665157,
+      "loss": 0.1326,
+      "step": 990
+    },
+    {
+      "epoch": 2.1645021645021645,
+      "grad_norm": 0.341796875,
+      "learning_rate": 0.00029638009049773756,
+      "loss": 0.1335,
+      "step": 1000
+    },
+    {
+      "epoch": 2.1645021645021645,
+      "eval_loss": 0.11199549585580826,
+      "eval_runtime": 20.4528,
+      "eval_samples_per_second": 24.447,
+      "eval_steps_per_second": 0.782,
+      "step": 1000
+    },
+    {
+      "epoch": 2.186147186147186,
+      "grad_norm": 0.369140625,
+      "learning_rate": 0.00029411764705882356,
+      "loss": 0.1375,
+      "step": 1010
+    },
+    {
+      "epoch": 2.207792207792208,
+      "grad_norm": 0.271484375,
+      "learning_rate": 0.00029185520361990955,
+      "loss": 0.1334,
+      "step": 1020
+    },
+    {
+      "epoch": 2.2294372294372296,
+      "grad_norm": 0.294921875,
+      "learning_rate": 0.0002895927601809955,
+      "loss": 0.1332,
+      "step": 1030
+    },
+    {
+      "epoch": 2.2510822510822512,
+      "grad_norm": 0.392578125,
+      "learning_rate": 0.00028733031674208143,
+      "loss": 0.1327,
+      "step": 1040
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.271484375,
+      "learning_rate": 0.0002850678733031674,
+      "loss": 0.141,
+      "step": 1050
+    },
+    {
+      "epoch": 2.2943722943722946,
+      "grad_norm": 0.357421875,
+      "learning_rate": 0.0002828054298642534,
+      "loss": 0.1432,
+      "step": 1060
+    },
+    {
+      "epoch": 2.316017316017316,
+      "grad_norm": 0.2890625,
+      "learning_rate": 0.00028054298642533936,
+      "loss": 0.1447,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3376623376623376,
+      "grad_norm": 0.2890625,
+      "learning_rate": 0.00027828054298642535,
+      "loss": 0.1376,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3593073593073592,
+      "grad_norm": 0.29296875,
+      "learning_rate": 0.00027601809954751135,
+      "loss": 0.1287,
+      "step": 1090
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.333984375,
+      "learning_rate": 0.00027375565610859734,
+      "loss": 0.1399,
+      "step": 1100
+    },
+    {
+      "epoch": 2.380952380952381,
+      "eval_loss": 0.09553591907024384,
+      "eval_runtime": 20.0139,
+      "eval_samples_per_second": 24.983,
+      "eval_steps_per_second": 0.799,
+      "step": 1100
+    },
+    {
+      "epoch": 2.4025974025974026,
+      "grad_norm": 0.318359375,
+      "learning_rate": 0.0002714932126696832,
+      "loss": 0.1417,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 0.294921875,
+      "learning_rate": 0.0002692307692307692,
+      "loss": 0.1428,
+      "step": 1120
+    },
+    {
+      "epoch": 2.445887445887446,
+      "grad_norm": 0.30078125,
+      "learning_rate": 0.0002669683257918552,
+      "loss": 0.1367,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4675324675324677,
+      "grad_norm": 0.271484375,
+      "learning_rate": 0.0002647058823529412,
+      "loss": 0.1383,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4891774891774894,
+      "grad_norm": 0.2578125,
+      "learning_rate": 0.00026244343891402715,
+      "loss": 0.1438,
+      "step": 1150
+    },
+    {
+      "epoch": 2.5108225108225106,
+      "grad_norm": 0.3515625,
+      "learning_rate": 0.00026018099547511314,
+      "loss": 0.1394,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5324675324675323,
+      "grad_norm": 0.291015625,
+      "learning_rate": 0.00025791855203619913,
+      "loss": 0.1358,
+      "step": 1170
+    },
+    {
+      "epoch": 2.554112554112554,
+      "grad_norm": 0.326171875,
+      "learning_rate": 0.0002556561085972851,
+      "loss": 0.1266,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5757575757575757,
+      "grad_norm": 0.3828125,
+      "learning_rate": 0.000253393665158371,
+      "loss": 0.1353,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5974025974025974,
+      "grad_norm": 0.30078125,
+      "learning_rate": 0.000251131221719457,
+      "loss": 0.1483,
+      "step": 1200
+    },
+    {
+      "epoch": 2.5974025974025974,
+      "eval_loss": 0.10388709604740143,
+      "eval_runtime": 19.843,
+      "eval_samples_per_second": 25.198,
+      "eval_steps_per_second": 0.806,
+      "step": 1200
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.2490234375,
+      "learning_rate": 0.000248868778280543,
+      "loss": 0.1364,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6406926406926408,
+      "grad_norm": 0.2080078125,
+      "learning_rate": 0.000246606334841629,
+      "loss": 0.1285,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6623376623376624,
+      "grad_norm": 0.251953125,
+      "learning_rate": 0.00024434389140271494,
+      "loss": 0.1264,
+      "step": 1230
+    },
+    {
+      "epoch": 2.683982683982684,
+      "grad_norm": 0.287109375,
+      "learning_rate": 0.0002420814479638009,
+      "loss": 0.1331,
+      "step": 1240
+    },
+    {
+      "epoch": 2.7056277056277054,
+      "grad_norm": 0.396484375,
+      "learning_rate": 0.0002398190045248869,
+      "loss": 0.15,
+      "step": 1250
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.275390625,
+      "learning_rate": 0.00023755656108597284,
+      "loss": 0.1367,
+      "step": 1260
+    },
+    {
+      "epoch": 2.7489177489177488,
+      "grad_norm": 0.275390625,
+      "learning_rate": 0.00023529411764705883,
+      "loss": 0.13,
+      "step": 1270
+    },
+    {
+      "epoch": 2.7705627705627704,
+      "grad_norm": 0.33203125,
+      "learning_rate": 0.0002330316742081448,
+      "loss": 0.1324,
+      "step": 1280
+    },
+    {
+      "epoch": 2.792207792207792,
+      "grad_norm": 0.302734375,
+      "learning_rate": 0.0002307692307692308,
+      "loss": 0.1336,
+      "step": 1290
+    },
+    {
+      "epoch": 2.813852813852814,
+      "grad_norm": 0.23828125,
+      "learning_rate": 0.00022850678733031673,
+      "loss": 0.1314,
+      "step": 1300
+    },
+    {
+      "epoch": 2.813852813852814,
+      "eval_loss": 0.10149528831243515,
+      "eval_runtime": 19.1293,
+      "eval_samples_per_second": 26.138,
+      "eval_steps_per_second": 0.836,
+      "step": 1300
+    },
+    {
+      "epoch": 2.8354978354978355,
+      "grad_norm": 0.306640625,
+      "learning_rate": 0.00022624434389140272,
+      "loss": 0.1309,
+      "step": 1310
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.2734375,
+      "learning_rate": 0.0002239819004524887,
+      "loss": 0.1353,
+      "step": 1320
+    },
+    {
+      "epoch": 2.878787878787879,
+      "grad_norm": 0.462890625,
+      "learning_rate": 0.00022171945701357466,
+      "loss": 0.1446,
+      "step": 1330
+    },
+    {
+      "epoch": 2.9004329004329006,
+      "grad_norm": 0.26953125,
+      "learning_rate": 0.00021945701357466062,
+      "loss": 0.1333,
+      "step": 1340
+    },
+    {
+      "epoch": 2.9220779220779223,
+      "grad_norm": 0.2080078125,
+      "learning_rate": 0.00021719457013574662,
+      "loss": 0.1383,
+      "step": 1350
+    },
+    {
+      "epoch": 2.9437229437229435,
+      "grad_norm": 0.267578125,
+      "learning_rate": 0.00021493212669683259,
+      "loss": 0.1496,
+      "step": 1360
+    },
+    {
+      "epoch": 2.965367965367965,
+      "grad_norm": 0.283203125,
+      "learning_rate": 0.00021266968325791855,
+      "loss": 0.1399,
+      "step": 1370
+    },
+    {
+      "epoch": 2.987012987012987,
+      "grad_norm": 0.380859375,
+      "learning_rate": 0.00021040723981900452,
+      "loss": 0.139,
+      "step": 1380
+    },
+    {
+      "epoch": 3.0086580086580086,
+      "grad_norm": 0.201171875,
+      "learning_rate": 0.0002081447963800905,
+      "loss": 0.127,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0303030303030303,
+      "grad_norm": 0.2578125,
+      "learning_rate": 0.00020588235294117645,
+      "loss": 0.114,
+      "step": 1400
+    },
+    {
+      "epoch": 3.0303030303030303,
+      "eval_loss": 0.09961362928152084,
+      "eval_runtime": 20.1151,
+      "eval_samples_per_second": 24.857,
+      "eval_steps_per_second": 0.795,
+      "step": 1400
+    },
+    {
+      "epoch": 3.051948051948052,
+      "grad_norm": 0.298828125,
+      "learning_rate": 0.00020361990950226245,
+      "loss": 0.1222,
+      "step": 1410
+    },
+    {
+      "epoch": 3.0735930735930737,
+      "grad_norm": 0.2490234375,
+      "learning_rate": 0.0002013574660633484,
+      "loss": 0.1174,
+      "step": 1420
+    },
+    {
+      "epoch": 3.0952380952380953,
+      "grad_norm": 0.2275390625,
+      "learning_rate": 0.00019909502262443438,
+      "loss": 0.121,
+      "step": 1430
+    },
+    {
+      "epoch": 3.116883116883117,
+      "grad_norm": 0.33203125,
+      "learning_rate": 0.00019683257918552037,
+      "loss": 0.1179,
+      "step": 1440
+    },
+    {
+      "epoch": 3.1385281385281387,
+      "grad_norm": 0.2373046875,
+      "learning_rate": 0.00019457013574660634,
+      "loss": 0.1163,
+      "step": 1450
+    },
+    {
+      "epoch": 3.16017316017316,
+      "grad_norm": 0.232421875,
+      "learning_rate": 0.00019230769230769233,
+      "loss": 0.1173,
+      "step": 1460
+    },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.2734375,
+      "learning_rate": 0.00019004524886877827,
+      "loss": 0.1193,
+      "step": 1470
+    },
+    {
+      "epoch": 3.2034632034632033,
+      "grad_norm": 0.458984375,
+      "learning_rate": 0.00018778280542986427,
+      "loss": 0.1208,
+      "step": 1480
+    },
+    {
+      "epoch": 3.225108225108225,
+      "grad_norm": 0.279296875,
+      "learning_rate": 0.00018552036199095024,
+      "loss": 0.1229,
+      "step": 1490
+    },
+    {
+      "epoch": 3.2467532467532467,
+      "grad_norm": 0.283203125,
+      "learning_rate": 0.0001832579185520362,
+      "loss": 0.1202,
+      "step": 1500
+    },
+    {
+      "epoch": 3.2467532467532467,
+      "eval_loss": 0.10085491091012955,
+      "eval_runtime": 21.5392,
+      "eval_samples_per_second": 23.214,
+      "eval_steps_per_second": 0.743,
+      "step": 1500
+    },
+    {
+      "epoch": 3.2683982683982684,
+      "grad_norm": 0.1943359375,
+      "learning_rate": 0.00018099547511312217,
+      "loss": 0.119,
+      "step": 1510
+    },
+    {
+      "epoch": 3.29004329004329,
+      "grad_norm": 0.45703125,
+      "learning_rate": 0.00017873303167420816,
+      "loss": 0.124,
+      "step": 1520
+    },
+    {
+      "epoch": 3.311688311688312,
+      "grad_norm": 0.296875,
+      "learning_rate": 0.00017647058823529413,
+      "loss": 0.1214,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.326171875,
+      "learning_rate": 0.0001742081447963801,
+      "loss": 0.1236,
+      "step": 1540
+    },
+    {
+      "epoch": 3.354978354978355,
+      "grad_norm": 0.369140625,
+      "learning_rate": 0.00017194570135746606,
+      "loss": 0.1175,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3766233766233764,
+      "grad_norm": 0.2236328125,
+      "learning_rate": 0.00016968325791855206,
+      "loss": 0.1163,
+      "step": 1560
+    },
+    {
+      "epoch": 3.398268398268398,
+      "grad_norm": 0.361328125,
+      "learning_rate": 0.000167420814479638,
+      "loss": 0.1114,
+      "step": 1570
+    },
+    {
+      "epoch": 3.41991341991342,
+      "grad_norm": 0.1845703125,
+      "learning_rate": 0.000165158371040724,
+      "loss": 0.1164,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4415584415584415,
+      "grad_norm": 0.2060546875,
+      "learning_rate": 0.00016289592760180996,
+      "loss": 0.1128,
+      "step": 1590
+    },
+    {
+      "epoch": 3.463203463203463,
+      "grad_norm": 0.3125,
+      "learning_rate": 0.00016063348416289595,
+      "loss": 0.1163,
+      "step": 1600
+    },
+    {
+      "epoch": 3.463203463203463,
+      "eval_loss": 0.10191706568002701,
+      "eval_runtime": 23.6847,
+      "eval_samples_per_second": 21.111,
+      "eval_steps_per_second": 0.676,
+      "step": 1600
+    },
+    {
+      "epoch": 3.484848484848485,
+      "grad_norm": 0.287109375,
+      "learning_rate": 0.0001583710407239819,
+      "loss": 0.1171,
+      "step": 1610
+    },
+    {
+      "epoch": 3.5064935064935066,
+      "grad_norm": 0.310546875,
+      "learning_rate": 0.00015610859728506788,
+      "loss": 0.1159,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5281385281385282,
+      "grad_norm": 0.2451171875,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 0.12,
+      "step": 1630
+    },
+    {
+      "epoch": 3.54978354978355,
+      "grad_norm": 0.208984375,
+      "learning_rate": 0.00015158371040723982,
+      "loss": 0.1048,
+      "step": 1640
+    },
+    {
+      "epoch": 3.571428571428571,
+      "grad_norm": 0.1591796875,
+      "learning_rate": 0.00014932126696832579,
+      "loss": 0.1061,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5930735930735933,
+      "grad_norm": 0.416015625,
+      "learning_rate": 0.00014705882352941178,
+      "loss": 0.1283,
+      "step": 1660
+    },
+    {
+      "epoch": 3.6147186147186146,
+      "grad_norm": 0.248046875,
+      "learning_rate": 0.00014479638009049775,
+      "loss": 0.1194,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.228515625,
+      "learning_rate": 0.0001425339366515837,
+      "loss": 0.1132,
+      "step": 1680
+    },
+    {
+      "epoch": 3.658008658008658,
+      "grad_norm": 0.361328125,
+      "learning_rate": 0.00014027149321266968,
+      "loss": 0.1226,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6796536796536796,
+      "grad_norm": 0.32421875,
+      "learning_rate": 0.00013800904977375567,
+      "loss": 0.1245,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6796536796536796,
+      "eval_loss": 0.09824839979410172,
+      "eval_runtime": 19.2671,
+      "eval_samples_per_second": 25.951,
+      "eval_steps_per_second": 0.83,
+      "step": 1700
+    },
+    {
+      "epoch": 3.7012987012987013,
+      "grad_norm": 0.365234375,
+      "learning_rate": 0.0001357466063348416,
+      "loss": 0.1219,
+      "step": 1710
+    },
+    {
+      "epoch": 3.722943722943723,
+      "grad_norm": 0.365234375,
+      "learning_rate": 0.0001334841628959276,
+      "loss": 0.1184,
+      "step": 1720
+    },
+    {
+      "epoch": 3.7445887445887447,
+      "grad_norm": 0.2451171875,
+      "learning_rate": 0.00013122171945701357,
+      "loss": 0.1157,
+      "step": 1730
+    },
+    {
+      "epoch": 3.7662337662337664,
+      "grad_norm": 0.263671875,
+      "learning_rate": 0.00012895927601809957,
+      "loss": 0.1075,
+      "step": 1740
+    },
+    {
+      "epoch": 3.787878787878788,
+      "grad_norm": 0.265625,
+      "learning_rate": 0.0001266968325791855,
+      "loss": 0.1102,
+      "step": 1750
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.28125,
+      "learning_rate": 0.0001244343891402715,
+      "loss": 0.1184,
+      "step": 1760
+    },
+    {
+      "epoch": 3.8311688311688314,
+      "grad_norm": 0.27734375,
+      "learning_rate": 0.00012217194570135747,
+      "loss": 0.1081,
+      "step": 1770
+    },
+    {
+      "epoch": 3.8528138528138527,
+      "grad_norm": 0.365234375,
+      "learning_rate": 0.00011990950226244345,
+      "loss": 0.1222,
+      "step": 1780
+    },
+    {
+      "epoch": 3.8744588744588744,
+      "grad_norm": 0.328125,
+      "learning_rate": 0.00011764705882352942,
+      "loss": 0.1224,
+      "step": 1790
+    },
+    {
+      "epoch": 3.896103896103896,
+      "grad_norm": 0.228515625,
+      "learning_rate": 0.0001153846153846154,
+      "loss": 0.1032,
+      "step": 1800
+    },
+    {
+      "epoch": 3.896103896103896,
+      "eval_loss": 0.09573096036911011,
+      "eval_runtime": 23.7902,
+      "eval_samples_per_second": 21.017,
+      "eval_steps_per_second": 0.673,
+      "step": 1800
+    },
+    {
+      "epoch": 3.9177489177489178,
+      "grad_norm": 0.33984375,
+      "learning_rate": 0.00011312217194570136,
+      "loss": 0.1182,
+      "step": 1810
+    },
+    {
+      "epoch": 3.9393939393939394,
+      "grad_norm": 0.3828125,
+      "learning_rate": 0.00011085972850678733,
+      "loss": 0.1191,
+      "step": 1820
+    },
+    {
+      "epoch": 3.961038961038961,
+      "grad_norm": 0.3359375,
+      "learning_rate": 0.00010859728506787331,
+      "loss": 0.1125,
+      "step": 1830
+    },
+    {
+      "epoch": 3.982683982683983,
+      "grad_norm": 0.2138671875,
+      "learning_rate": 0.00010633484162895928,
+      "loss": 0.1188,
+      "step": 1840
+    },
+    {
+      "epoch": 4.004329004329004,
+      "grad_norm": 0.30078125,
+      "learning_rate": 0.00010407239819004526,
+      "loss": 0.1203,
+      "step": 1850
+    },
+    {
+      "epoch": 4.025974025974026,
+      "grad_norm": 0.232421875,
+      "learning_rate": 0.00010180995475113122,
+      "loss": 0.1077,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0476190476190474,
+      "grad_norm": 0.2080078125,
+      "learning_rate": 9.954751131221719e-05,
+      "loss": 0.1065,
+      "step": 1870
+    },
+    {
+      "epoch": 4.06926406926407,
+      "grad_norm": 0.27734375,
+      "learning_rate": 9.728506787330317e-05,
+      "loss": 0.1112,
+      "step": 1880
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 0.2333984375,
+      "learning_rate": 9.502262443438914e-05,
+      "loss": 0.1006,
+      "step": 1890
+    },
+    {
+      "epoch": 4.112554112554113,
+      "grad_norm": 0.205078125,
+      "learning_rate": 9.276018099547512e-05,
+      "loss": 0.1043,
+      "step": 1900
+    },
+    {
+      "epoch": 4.112554112554113,
+      "eval_loss": 0.09408007562160492,
+      "eval_runtime": 19.9413,
+      "eval_samples_per_second": 25.074,
+      "eval_steps_per_second": 0.802,
+      "step": 1900
+    },
+    {
+      "epoch": 4.134199134199134,
+      "grad_norm": 0.265625,
+      "learning_rate": 9.049773755656108e-05,
+      "loss": 0.1013,
+      "step": 1910
+    },
+    {
+      "epoch": 4.1558441558441555,
+      "grad_norm": 0.2109375,
+      "learning_rate": 8.823529411764706e-05,
+      "loss": 0.1058,
+      "step": 1920
+    },
+    {
+      "epoch": 4.177489177489178,
+      "grad_norm": 0.19140625,
+      "learning_rate": 8.597285067873303e-05,
+      "loss": 0.1069,
+      "step": 1930
+    },
+    {
+      "epoch": 4.199134199134199,
+      "grad_norm": 0.267578125,
+      "learning_rate": 8.3710407239819e-05,
+      "loss": 0.1089,
+      "step": 1940
+    },
+    {
+      "epoch": 4.220779220779221,
+      "grad_norm": 0.283203125,
+      "learning_rate": 8.144796380090498e-05,
+      "loss": 0.1026,
+      "step": 1950
+    },
+    {
+      "epoch": 4.242424242424242,
+      "grad_norm": 0.357421875,
+      "learning_rate": 7.918552036199095e-05,
+      "loss": 0.1043,
+      "step": 1960
+    },
+    {
+      "epoch": 4.264069264069264,
+      "grad_norm": 0.21484375,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 0.1136,
+      "step": 1970
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.3203125,
+      "learning_rate": 7.466063348416289e-05,
+      "loss": 0.1045,
+      "step": 1980
+    },
+    {
+      "epoch": 4.307359307359308,
+      "grad_norm": 0.244140625,
+      "learning_rate": 7.239819004524887e-05,
+      "loss": 0.1032,
+      "step": 1990
+    },
+    {
+      "epoch": 4.329004329004329,
+      "grad_norm": 0.19140625,
+      "learning_rate": 7.013574660633484e-05,
+      "loss": 0.0977,
+      "step": 2000
+    },
+    {
+      "epoch": 4.329004329004329,
+      "eval_loss": 0.09685727208852768,
+      "eval_runtime": 20.4475,
+      "eval_samples_per_second": 24.453,
+      "eval_steps_per_second": 0.782,
+      "step": 2000
+    },
+    {
+      "epoch": 4.35064935064935,
+      "grad_norm": 0.2333984375,
+      "learning_rate": 6.78733031674208e-05,
+      "loss": 0.1019,
+      "step": 2010
+    },
+    {
+      "epoch": 4.372294372294372,
+      "grad_norm": 0.2294921875,
+      "learning_rate": 6.561085972850679e-05,
+      "loss": 0.1076,
+      "step": 2020
+    },
+    {
+      "epoch": 4.393939393939394,
+      "grad_norm": 0.1650390625,
+      "learning_rate": 6.334841628959275e-05,
+      "loss": 0.0983,
+      "step": 2030
+    },
+    {
+      "epoch": 4.415584415584416,
+      "grad_norm": 0.21484375,
+      "learning_rate": 6.108597285067873e-05,
+      "loss": 0.1023,
+      "step": 2040
+    },
+    {
+      "epoch": 4.437229437229437,
+      "grad_norm": 0.26953125,
+      "learning_rate": 5.882352941176471e-05,
+      "loss": 0.0982,
+      "step": 2050
+    },
+    {
+      "epoch": 4.458874458874459,
+      "grad_norm": 0.306640625,
+      "learning_rate": 5.656108597285068e-05,
+      "loss": 0.1071,
+      "step": 2060
+    },
+    {
+      "epoch": 4.48051948051948,
+      "grad_norm": 0.2265625,
+      "learning_rate": 5.4298642533936655e-05,
+      "loss": 0.1095,
+      "step": 2070
+    },
+    {
+      "epoch": 4.5021645021645025,
+      "grad_norm": 0.240234375,
+      "learning_rate": 5.203619909502263e-05,
+      "loss": 0.1105,
+      "step": 2080
+    },
+    {
+      "epoch": 4.523809523809524,
+      "grad_norm": 0.2734375,
+      "learning_rate": 4.9773755656108595e-05,
+      "loss": 0.0961,
+      "step": 2090
+    },
+    {
+      "epoch": 4.545454545454545,
+      "grad_norm": 0.2412109375,
+      "learning_rate": 4.751131221719457e-05,
+      "loss": 0.1016,
+      "step": 2100
+    },
+    {
+      "epoch": 4.545454545454545,
+      "eval_loss": 0.09792981296777725,
+      "eval_runtime": 20.8232,
+      "eval_samples_per_second": 24.012,
+      "eval_steps_per_second": 0.768,
+      "step": 2100
+    },
+    {
+      "epoch": 4.567099567099567,
+      "grad_norm": 0.3125,
+      "learning_rate": 4.524886877828054e-05,
+      "loss": 0.0953,
+      "step": 2110
+    },
+    {
+      "epoch": 4.588744588744589,
+      "grad_norm": 0.2421875,
+      "learning_rate": 4.2986425339366516e-05,
+      "loss": 0.105,
+      "step": 2120
+    },
+    {
+      "epoch": 4.6103896103896105,
+      "grad_norm": 0.255859375,
+      "learning_rate": 4.072398190045249e-05,
+      "loss": 0.1063,
+      "step": 2130
+    },
+    {
+      "epoch": 4.632034632034632,
+      "grad_norm": 0.33203125,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.0982,
+      "step": 2140
+    },
+    {
+      "epoch": 4.653679653679654,
+      "grad_norm": 0.265625,
+      "learning_rate": 3.6199095022624436e-05,
+      "loss": 0.1068,
+      "step": 2150
+    },
+    {
+      "epoch": 4.675324675324675,
+      "grad_norm": 0.2421875,
+      "learning_rate": 3.39366515837104e-05,
+      "loss": 0.1111,
+      "step": 2160
+    },
+    {
+      "epoch": 4.696969696969697,
+      "grad_norm": 0.189453125,
+      "learning_rate": 3.167420814479638e-05,
+      "loss": 0.1084,
+      "step": 2170
+    },
+    {
+      "epoch": 4.7186147186147185,
+      "grad_norm": 0.2734375,
+      "learning_rate": 2.9411764705882354e-05,
+      "loss": 0.103,
+      "step": 2180
+    },
+    {
+      "epoch": 4.740259740259741,
+      "grad_norm": 0.337890625,
+      "learning_rate": 2.7149321266968327e-05,
+      "loss": 0.1089,
+      "step": 2190
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.3046875,
+      "learning_rate": 2.4886877828054298e-05,
+      "loss": 0.097,
+      "step": 2200
+    },
+    {
+      "epoch": 4.761904761904762,
+      "eval_loss": 0.09769493341445923,
+      "eval_runtime": 19.6594,
+      "eval_samples_per_second": 25.433,
+      "eval_steps_per_second": 0.814,
+      "step": 2200
+    },
+    {
+      "epoch": 4.783549783549784,
+      "grad_norm": 0.296875,
+      "learning_rate": 2.262443438914027e-05,
+      "loss": 0.1083,
+      "step": 2210
+    },
+    {
+      "epoch": 4.805194805194805,
+      "grad_norm": 0.2158203125,
+      "learning_rate": 2.0361990950226245e-05,
+      "loss": 0.1046,
+      "step": 2220
+    },
+    {
+      "epoch": 4.8268398268398265,
+      "grad_norm": 0.26953125,
+      "learning_rate": 1.8099547511312218e-05,
+      "loss": 0.1139,
+      "step": 2230
+    },
+    {
+      "epoch": 4.848484848484849,
+      "grad_norm": 0.28515625,
+      "learning_rate": 1.583710407239819e-05,
+      "loss": 0.1038,
+      "step": 2240
+    },
+    {
+      "epoch": 4.87012987012987,
+      "grad_norm": 0.298828125,
+      "learning_rate": 1.3574660633484164e-05,
+      "loss": 0.1089,
+      "step": 2250
+    },
+    {
+      "epoch": 4.891774891774892,
+      "grad_norm": 0.23046875,
+      "learning_rate": 1.1312217194570136e-05,
+      "loss": 0.0922,
+      "step": 2260
+    },
+    {
+      "epoch": 4.913419913419913,
+      "grad_norm": 0.2578125,
+      "learning_rate": 9.049773755656109e-06,
+      "loss": 0.1041,
+      "step": 2270
+    },
+    {
+      "epoch": 4.935064935064935,
+      "grad_norm": 0.255859375,
+      "learning_rate": 6.787330316742082e-06,
+      "loss": 0.0933,
+      "step": 2280
+    },
+    {
+      "epoch": 4.956709956709957,
+      "grad_norm": 0.29296875,
+      "learning_rate": 4.5248868778280546e-06,
+      "loss": 0.1089,
+      "step": 2290
+    },
+    {
+      "epoch": 4.978354978354979,
+      "grad_norm": 0.296875,
+      "learning_rate": 2.2624434389140273e-06,
+      "loss": 0.1094,
+      "step": 2300
+    },
+    {
+      "epoch": 4.978354978354979,
+      "eval_loss": 0.09713861346244812,
+      "eval_runtime": 20.5975,
+      "eval_samples_per_second": 24.275,
+      "eval_steps_per_second": 0.777,
+      "step": 2300
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.3359375,
+      "learning_rate": 0.0,
+      "loss": 0.1107,
+      "step": 2310
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2310,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.556491100242671e+18,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2310/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e723165fb4ce59d86427d7b00f9accdcf7438e283d47a4bdb986c5618c13fa4
+size 5176

git_hash.txt

@@ -0,0 +1 @@
+961d20526206465f9ebef9affd131d53a9f516e6

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

results.json

@@ -0,0 +1 @@
+{"../colpali/data_dir/eval_vidore/tatdqa_test": {"ndcg_at_1": 0.70109, "ndcg_at_3": 0.7923, "ndcg_at_5": 0.81432, "ndcg_at_10": 0.82714, "ndcg_at_20": 0.83121, "ndcg_at_50": 0.83523, "ndcg_at_100": 0.83688, "map_at_1": 0.70109, "map_at_3": 0.77005, "map_at_5": 0.78241, "map_at_10": 0.78785, "map_at_20": 0.78901, "map_at_50": 0.78967, "map_at_100": 0.78981, "recall_at_1": 0.70109, "recall_at_3": 0.85662, "recall_at_5": 0.90948, "recall_at_10": 0.94836, "recall_at_20": 0.96416, "recall_at_50": 0.9842, "recall_at_100": 0.99453, "precision_at_1": 0.70109, "precision_at_3": 0.28554, "precision_at_5": 0.1819, "precision_at_10": 0.09484, "precision_at_20": 0.04821, "precision_at_50": 0.01968, "precision_at_100": 0.00995, "mrr_at_1": 0.7017010935601458, "mrr_at_3": 0.7709599027946537, "mrr_at_5": 0.7835965978128797, "mrr_at_10": 0.7887963123686088, "mrr_at_20": 0.7900069324299046, "mrr_at_50": 0.7906871013277383, "mrr_at_100": 0.7908122337289102, "naucs_at_1_max": 0.31323987398365244, "naucs_at_1_std": -0.06973458951296707, "naucs_at_1_diff1": 0.8257166838058789, "naucs_at_3_max": 0.3926659737473996, "naucs_at_3_std": 0.06571917292636736, "naucs_at_3_diff1": 0.7756394870292411, "naucs_at_5_max": 0.397380358594201, "naucs_at_5_std": 0.13868537387746546, "naucs_at_5_diff1": 0.7396051349982516, "naucs_at_10_max": 0.501090641931301, "naucs_at_10_std": 0.36694999105347903, "naucs_at_10_diff1": 0.7174193926609567, "naucs_at_20_max": 0.4765529924218107, "naucs_at_20_std": 0.3737660736114713, "naucs_at_20_diff1": 0.6900626633277733, "naucs_at_50_max": 0.6429809235278162, "naucs_at_50_std": 0.5965680225588305, "naucs_at_50_diff1": 0.7246754428536203, "naucs_at_100_max": 0.6364067564396122, "naucs_at_100_std": 0.5448683368190432, "naucs_at_100_diff1": 0.6812946217425379}, "../colpali/data_dir/eval_vidore/shiftproject_test": {"ndcg_at_1": 0.81, "ndcg_at_3": 0.8894, "ndcg_at_5": 0.90663, "ndcg_at_10": 0.90952, "ndcg_at_20": 0.90952, "ndcg_at_50": 0.9115, "ndcg_at_100": 0.9115, "map_at_1": 0.81, "map_at_3": 0.87167, "map_at_5": 0.88167, "map_at_10": 0.88267, "map_at_20": 0.88267, "map_at_50": 0.88298, "map_at_100": 0.88298, "recall_at_1": 0.81, "recall_at_3": 0.94, "recall_at_5": 0.98, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.81, "precision_at_3": 0.31333, "precision_at_5": 0.196, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.82, "mrr_at_3": 0.8766666666666667, "mrr_at_5": 0.8866666666666667, "mrr_at_10": 0.8877777777777777, "mrr_at_20": 0.8877777777777777, "mrr_at_50": 0.8880902777777777, "mrr_at_100": 0.8880902777777777, "naucs_at_1_max": 0.0649588768400645, "naucs_at_1_std": -0.2740845513122743, "naucs_at_1_diff1": 0.6852637644716851, "naucs_at_3_max": 0.3546529723000309, "naucs_at_3_std": -0.11009959539371503, "naucs_at_3_diff1": 0.5325241207594116, "naucs_at_5_max": 0.722222222222224, "naucs_at_5_std": 0.3384687208216692, "naucs_at_5_diff1": 0.42250233426704475, "naucs_at_10_max": 0.7222222222222276, "naucs_at_10_std": 0.5541549953314738, "naucs_at_10_diff1": 0.7222222222222276, "naucs_at_20_max": 0.7222222222222276, "naucs_at_20_std": 0.5541549953314738, "naucs_at_20_diff1": 0.7222222222222276, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/syntheticDocQA_artificial_intelligence_test": {"ndcg_at_1": 0.99, "ndcg_at_3": 0.99, "ndcg_at_5": 0.99387, "ndcg_at_10": 0.99387, "ndcg_at_20": 0.99387, "ndcg_at_50": 0.99387, "ndcg_at_100": 0.99387, "map_at_1": 0.99, "map_at_3": 0.99, "map_at_5": 0.992, "map_at_10": 0.992, "map_at_20": 0.992, "map_at_50": 0.992, "map_at_100": 0.992, "recall_at_1": 0.99, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.99, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.99, "mrr_at_3": 0.99, "mrr_at_5": 0.992, "mrr_at_10": 0.992, "mrr_at_20": 0.992, "mrr_at_50": 0.992, "mrr_at_100": 0.992, "naucs_at_1_max": 0.8692810457516276, "naucs_at_1_std": -0.5634920634920657, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 0.8692810457516356, "naucs_at_3_std": -0.5634920634921204, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/syntheticDocQA_government_reports_test": {"ndcg_at_1": 0.91, "ndcg_at_3": 0.95417, "ndcg_at_5": 0.96278, "ndcg_at_10": 0.96278, "ndcg_at_20": 0.96278, "ndcg_at_50": 0.96278, "ndcg_at_100": 0.96278, "map_at_1": 0.91, "map_at_3": 0.945, "map_at_5": 0.95, "map_at_10": 0.95, "map_at_20": 0.95, "map_at_50": 0.95, "map_at_100": 0.95, "recall_at_1": 0.91, "recall_at_3": 0.98, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.91, "precision_at_3": 0.32667, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.92, "mrr_at_3": 0.9566666666666667, "mrr_at_5": 0.9566666666666667, "mrr_at_10": 0.9566666666666667, "mrr_at_20": 0.9566666666666667, "mrr_at_50": 0.9566666666666667, "mrr_at_100": 0.9566666666666667, "naucs_at_1_max": 0.6380329909741665, "naucs_at_1_std": 0.3872289656603365, "naucs_at_1_diff1": 0.9419026870007259, "naucs_at_3_max": 0.8611111111111119, "naucs_at_3_std": 0.5401493930905577, "naucs_at_3_diff1": 0.9346405228758099, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/arxivqa_test_subsampled": {"ndcg_at_1": 0.824, "ndcg_at_3": 0.86576, "ndcg_at_5": 0.88065, "ndcg_at_10": 0.88761, "ndcg_at_20": 0.89161, "ndcg_at_50": 0.89571, "ndcg_at_100": 0.89671, "map_at_1": 0.824, "map_at_3": 0.85533, "map_at_5": 0.86363, "map_at_10": 0.86642, "map_at_20": 0.86749, "map_at_50": 0.8682, "map_at_100": 0.8683, "recall_at_1": 0.824, "recall_at_3": 0.896, "recall_at_5": 0.932, "recall_at_10": 0.954, "recall_at_20": 0.97, "recall_at_50": 0.99, "recall_at_100": 0.996, "precision_at_1": 0.824, "precision_at_3": 0.29867, "precision_at_5": 0.1864, "precision_at_10": 0.0954, "precision_at_20": 0.0485, "precision_at_50": 0.0198, "precision_at_100": 0.00996, "mrr_at_1": 0.826, "mrr_at_3": 0.8566666666666666, "mrr_at_5": 0.8647666666666667, "mrr_at_10": 0.8674404761904762, "mrr_at_20": 0.868713893416525, "mrr_at_50": 0.8694424481678229, "mrr_at_100": 0.8695414304104477, "naucs_at_1_max": 0.7663278157610652, "naucs_at_1_std": -0.09768014573587591, "naucs_at_1_diff1": 0.9187263853904275, "naucs_at_3_max": 0.8138309391467555, "naucs_at_3_std": -0.023388435717895166, "naucs_at_3_diff1": 0.8748014870425184, "naucs_at_5_max": 0.8622919756137744, "naucs_at_5_std": 0.09893172955456826, "naucs_at_5_diff1": 0.8950952930191681, "naucs_at_10_max": 0.8887671010433149, "naucs_at_10_std": 0.18117971826411405, "naucs_at_10_diff1": 0.9102829537612158, "naucs_at_20_max": 0.9081854964207933, "naucs_at_20_std": 0.31092436974789345, "naucs_at_20_diff1": 0.9193899782135069, "naucs_at_50_max": 1.0, "naucs_at_50_std": 0.7428571428571219, "naucs_at_50_diff1": 0.9215686274509768, "naucs_at_100_max": 1.0, "naucs_at_100_std": 0.5613912231558791, "naucs_at_100_diff1": 0.9346405228758466}, "../colpali/data_dir/eval_vidore/docvqa_test_subsampled": {"ndcg_at_1": 0.52993, "ndcg_at_3": 0.58573, "ndcg_at_5": 0.60558, "ndcg_at_10": 0.62266, "ndcg_at_20": 0.63582, "ndcg_at_50": 0.64908, "ndcg_at_100": 0.6581, "map_at_1": 0.52993, "map_at_3": 0.57206, "map_at_5": 0.58293, "map_at_10": 0.58991, "map_at_20": 0.59366, "map_at_50": 0.59581, "map_at_100": 0.59662, "recall_at_1": 0.52993, "recall_at_3": 0.62528, "recall_at_5": 0.67406, "recall_at_10": 0.72727, "recall_at_20": 0.77827, "recall_at_50": 0.84479, "recall_at_100": 0.90022, "precision_at_1": 0.52993, "precision_at_3": 0.20843, "precision_at_5": 0.13481, "precision_at_10": 0.07273, "precision_at_20": 0.03891, "precision_at_50": 0.0169, "precision_at_100": 0.009, "mrr_at_1": 0.5232815964523282, "mrr_at_3": 0.5728011825572801, "mrr_at_5": 0.5826681448632669, "mrr_at_10": 0.5891273360785556, "mrr_at_20": 0.5925723253011941, "mrr_at_50": 0.5943296372892853, "mrr_at_100": 0.59515893309537, "naucs_at_1_max": 0.1489445922678588, "naucs_at_1_std": 0.5944168012150516, "naucs_at_1_diff1": 0.9184631698098695, "naucs_at_3_max": 0.01752251361569101, "naucs_at_3_std": 0.7203256897672715, "naucs_at_3_diff1": 0.8894280781702324, "naucs_at_5_max": -0.04548428009159777, "naucs_at_5_std": 0.7690453232976177, "naucs_at_5_diff1": 0.8644439246273121, "naucs_at_10_max": -0.13892914655317037, "naucs_at_10_std": 0.8207534993760452, "naucs_at_10_diff1": 0.8438993391008719, "naucs_at_20_max": -0.27349363756086625, "naucs_at_20_std": 0.8575543278988474, "naucs_at_20_diff1": 0.8389957258659689, "naucs_at_50_max": -0.43528705588098854, "naucs_at_50_std": 0.9002887869267794, "naucs_at_50_diff1": 0.8407727389585528, "naucs_at_100_max": -0.4873376707138829, "naucs_at_100_std": 0.8779309480426964, "naucs_at_100_diff1": 0.8126388341509826}, "../colpali/data_dir/eval_vidore/syntheticDocQA_healthcare_industry_test": {"ndcg_at_1": 0.96, "ndcg_at_3": 0.97762, "ndcg_at_5": 0.98149, "ndcg_at_10": 0.98149, "ndcg_at_20": 0.98149, "ndcg_at_50": 0.98149, "ndcg_at_100": 0.98149, "map_at_1": 0.96, "map_at_3": 0.97333, "map_at_5": 0.97533, "map_at_10": 0.97533, "map_at_20": 0.97533, "map_at_50": 0.97533, "map_at_100": 0.97533, "recall_at_1": 0.96, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.96, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.96, "mrr_at_3": 0.9733333333333333, "mrr_at_5": 0.9753333333333333, "mrr_at_10": 0.9753333333333333, "mrr_at_20": 0.9753333333333333, "mrr_at_50": 0.9753333333333333, "mrr_at_100": 0.9753333333333333, "naucs_at_1_max": 0.6785714285714297, "naucs_at_1_std": -0.529178338001864, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 1.0, "naucs_at_3_std": 0.35807656395889226, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/infovqa_test_subsampled": {"ndcg_at_1": 0.89069, "ndcg_at_3": 0.91896, "ndcg_at_5": 0.92636, "ndcg_at_10": 0.93041, "ndcg_at_20": 0.93306, "ndcg_at_50": 0.93549, "ndcg_at_100": 0.93583, "map_at_1": 0.89069, "map_at_3": 0.91194, "map_at_5": 0.91599, "map_at_10": 0.91774, "map_at_20": 0.91851, "map_at_50": 0.91891, "map_at_100": 0.91894, "recall_at_1": 0.89069, "recall_at_3": 0.93927, "recall_at_5": 0.95749, "recall_at_10": 0.96964, "recall_at_20": 0.97976, "recall_at_50": 0.9919, "recall_at_100": 0.99393, "precision_at_1": 0.89069, "precision_at_3": 0.31309, "precision_at_5": 0.1915, "precision_at_10": 0.09696, "precision_at_20": 0.04899, "precision_at_50": 0.01984, "precision_at_100": 0.00994, "mrr_at_1": 0.888663967611336, "mrr_at_3": 0.9102564102564102, "mrr_at_5": 0.9148110661268556, "mrr_at_10": 0.9165558126084441, "mrr_at_20": 0.9173312497615903, "mrr_at_50": 0.917719246204226, "mrr_at_100": 0.917752984395859, "naucs_at_1_max": 0.6452597879999299, "naucs_at_1_std": -0.1583967206391628, "naucs_at_1_diff1": 0.9382939669972173, "naucs_at_3_max": 0.6792175380169426, "naucs_at_3_std": -0.07352824072374452, "naucs_at_3_diff1": 0.9330460900047705, "naucs_at_5_max": 0.8390629595590892, "naucs_at_5_std": 0.22116664077349757, "naucs_at_5_diff1": 0.9626852988386374, "naucs_at_10_max": 0.8657206615100057, "naucs_at_10_std": 0.2744376241740944, "naucs_at_10_diff1": 0.9564661819784096, "naucs_at_20_max": 0.9183327616354467, "naucs_at_20_std": 0.6804227380555364, "naucs_at_20_diff1": 0.9477594183740975, "naucs_at_50_max": 0.8979159520443043, "naucs_at_50_std": 0.674633185026432, "naucs_at_50_diff1": 0.9346992729676393, "naucs_at_100_max": 0.8638879360590712, "naucs_at_100_std": 0.5661775800352391, "naucs_at_100_diff1": 0.9129323639568517}, "../colpali/data_dir/eval_vidore/syntheticDocQA_energy_test": {"ndcg_at_1": 0.94, "ndcg_at_3": 0.95893, "ndcg_at_5": 0.95893, "ndcg_at_10": 0.96542, "ndcg_at_20": 0.96542, "ndcg_at_50": 0.9676, "ndcg_at_100": 0.9676, "map_at_1": 0.94, "map_at_3": 0.955, "map_at_5": 0.955, "map_at_10": 0.95768, "map_at_20": 0.95768, "map_at_50": 0.95811, "map_at_100": 0.95811, "recall_at_1": 0.94, "recall_at_3": 0.97, "recall_at_5": 0.97, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.94, "precision_at_3": 0.32333, "precision_at_5": 0.194, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.95, "mrr_at_3": 0.96, "mrr_at_5": 0.96, "mrr_at_10": 0.9625396825396826, "mrr_at_20": 0.9630659983291562, "mrr_at_50": 0.9630659983291562, "mrr_at_100": 0.9630659983291562, "naucs_at_1_max": 0.06847183317771653, "naucs_at_1_std": -0.9000933706816028, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 0.6150015561780285, "naucs_at_3_std": -1.21708683473389, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 0.6150015561780299, "naucs_at_5_std": -1.2170868347338937, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 0.7222222222222276, "naucs_at_10_std": -1.7399626517273863, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 0.7222222222222276, "naucs_at_20_std": -1.7399626517273863, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/tabfquad_test_subsampled": {"ndcg_at_1": 0.84286, "ndcg_at_3": 0.88737, "ndcg_at_5": 0.89459, "ndcg_at_10": 0.89794, "ndcg_at_20": 0.90691, "ndcg_at_50": 0.90985, "ndcg_at_100": 0.91045, "map_at_1": 0.84286, "map_at_3": 0.87679, "map_at_5": 0.88071, "map_at_10": 0.88203, "map_at_20": 0.88445, "map_at_50": 0.88496, "map_at_100": 0.88502, "recall_at_1": 0.84286, "recall_at_3": 0.91786, "recall_at_5": 0.93571, "recall_at_10": 0.94643, "recall_at_20": 0.98214, "recall_at_50": 0.99643, "recall_at_100": 1.0, "precision_at_1": 0.84286, "precision_at_3": 0.30595, "precision_at_5": 0.18714, "precision_at_10": 0.09464, "precision_at_20": 0.04911, "precision_at_50": 0.01993, "precision_at_100": 0.01, "mrr_at_1": 0.8392857142857143, "mrr_at_3": 0.8744047619047619, "mrr_at_5": 0.8783333333333333, "mrr_at_10": 0.8797108843537415, "mrr_at_20": 0.8821667166283763, "mrr_at_50": 0.8826871907663073, "mrr_at_100": 0.8827487671209872, "naucs_at_1_max": 0.5016709740664825, "naucs_at_1_std": 0.19134881026646522, "naucs_at_1_diff1": 0.9055454059353004, "naucs_at_3_max": 0.6889741403807901, "naucs_at_3_std": 0.34459870904883405, "naucs_at_3_diff1": 0.8472780416514442, "naucs_at_5_max": 0.6857039111941077, "naucs_at_5_std": 0.40758896151053337, "naucs_at_5_diff1": 0.8275495383338527, "naucs_at_10_max": 0.6228446934329287, "naucs_at_10_std": 0.31633986928104846, "naucs_at_10_diff1": 0.8017740429505158, "naucs_at_20_max": 0.9183006535947714, "naucs_at_20_std": 0.7605042016806759, "naucs_at_20_diff1": 0.9183006535947714, "naucs_at_50_max": 1.0, "naucs_at_50_std": 1.0, "naucs_at_50_diff1": 0.8692810457515607, "naucs_at_100_max": 1.0, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 1.0}}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
+size 11420371

tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "ColQwen2Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_config.yml

@@ -0,0 +1,73 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2-ba256-ckpt-5e-nopad
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2Processor
+    pretrained_model_name_or_path:  "./models/colqwen2_base" # "./models/paligemma-3b-mix-448"
+    # num_image_tokens: 2048
+    # max_length: 50
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2
+    pretrained_model_name_or_path: "./models/colqwen2_base"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+#    device_map: "auto"
+#    quantization_config:
+#      (): transformers.BitsAndBytesConfig
+#      load_in_4bit: true
+#      bnb_4bit_quant_type: "nf4"
+#      bnb_4bit_compute_dtype:  "bfloat16"
+#      bnb_4bit_use_double_quant: true
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  add_suffix: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 5
+    per_device_train_batch_size: 64
+    gradient_checkpointing: true
+    gradient_checkpointing_kwargs: { "use_reentrant": false }
+    # gradient_checkpointing: true
+    # 6 x 8 gpus = 48 batch size
+    # gradient_accumulation_steps: 4
+    per_device_eval_batch_size: 8
+    eval_strategy: "steps"
+    dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 500
+    logging_steps: 10
+    eval_steps: 100
+    warmup_steps: 100
+    learning_rate: 5e-4
+    save_total_limit: 1
+    # resume_from_checkpoint: true
+    # optim: "paged_adamw_8bit"
+    # wandb logging
+    # wandb_project: "colqwen2"
+    # run_name: "colqwen2-ba32-nolora"
+    report_to: "wandb"
+
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 32
+    lora_alpha: 32
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+